Title: Modular design of underground rescue drilling rigs based on the fusion of knowledge and genetic clustering

Authors: Dong Fan

Addresses: CCTEG Xi'an Research Institute (Group) Co., Ltd., Xi'an, 710077, China; Faculty of Engineering, China University of Geosciences, Wuhan, 430074, China

Abstract: To address transportation difficulties and low assembly efficiency of large-diameter casing drilling rigs in underground coal mine rescue, this study proposes a knowledge-driven modular design method, integrating domain design knowledge (DSM), algorithm optimisation knowledge (MDL), and engineering evaluation knowledge (adaptability evaluation) to establish a systematic solution. A quantifiable design structure matrix was established based on a three-dimensional knowledge weighting model (structural: 0.68, geometric: 0.08, functional: 0.24). Combining MDL theory with K-means and genetic algorithms achieved optimal knowledge encapsulation, yielding seven module schemes (MDL: 522.17). Developed core modules including a dual-rotation high-torque power head (250 kNm) and independently-driven feed unit (1,604 kN) through multi-source knowledge integration. Knowledge-enhanced entropy-weighted TOPSIS evaluation showed outstanding performance (technical: 85 ± 2.3; economic: 92 ± 1.7; overall: 89.21). Field results demonstrated 60% faster assembly, 40% lower maintenance costs, and compliance with 4,000 × 1,750 × 1,800 mm roadway limits, establishing a smart equipment design standard for mine rescue.

Keywords: knowledge-driven; GA-optimised clustering method; underground rescue drill rig; modular design; design structure matrix; DSM.

DOI: 10.1504/IJBIC.2026.155170

International Journal of Bio-Inspired Computation, 2026 Vol.27 No.4, pp.203 - 219

Received: 07 Aug 2025
Accepted: 13 Nov 2025
Published online: 28 Jul 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article